From: Tiwei Bie <tiwei.bie@intel.com>
To: virtio-dev@lists.oasis-open.org
Cc: mst@redhat.com, cohuck@redhat.com, stefanha@redhat.com
Subject: [virtio-dev] [PATCH v4] packed-ring: fix used descriptor checking in example code
Date: Thu, 6 Dec 2018 00:27:01 +0800 [thread overview]
Message-ID: <20181205162701.14138-1-tiwei.bie@intel.com> (raw)
When the driver is processing used descriptors in parallel
with adding new available descriptors, the driver can't just
check whether USED bit equals to the used wrap counter when
checking whether a descriptor is a used descriptor, because
the driver also needs to check whether the descriptor has
been made available. Below is an example:
Assuming ring size is 4, ring's initial state will be:
+----+----+----+----+
| 00 | 00 | 00 | 00 |
+----+----+----+----+
00 means AVAIL=0 USED=0, 01 means AVAIL=0 USED=1
10 means AVAIL=1 USED=0, 11 means AVAIL=1 USED=1
After the driver made two descriptor chains available and
each chain consists of two descriptors, the ring could be:
+----+-----------+----+-----------+
| 10 | 10 (id=0) | 10 | 10 (id=1) |
+----+-----------+----+-----------+
After the device processed all the available descriptors and
made them used (e.g. in order), the ring could be:
+-----------+----+-----------+----+
| 11 (id=0) | 10 | 11 (id=1) | 10 |
+-----------+----+-----------+----+
After the driver processed all the used descriptors and made
one descriptor (not chained, just one descriptor) available,
the ring could be:
+-----------+----+----+----+
| 01 (id=0) | 10 | 11 | 10 |
+-----------+----+----+----+
After the device made that descriptor used, the ring will be:
+-----------+----+----+----+
| 00 (id=0) | 10 | 11 | 10 |
+-----------+----+----+----+
If the driver just checks whether USED bit equals to the used
wrap counter when checking whether a descriptor is a used
descriptor, after processing the first descriptor (whose
AVAIL and USED bits are both 0), and advancing vq->next_used
pointer, it will then also treat the next descriptor, i.e.
the second descriptor (whose AVAIL and USED bits are 1 and
0 respectively) as a used descriptor which is wrong.
Fixes: https://github.com/oasis-tcs/virtio-spec/issues/29
Signed-off-by: Tiwei Bie <tiwei.bie@intel.com>
---
v2:
- Add "Fixes" tag;
- Refine commit log;
v3:
- Compare with vq->used_wrap_count (MST);
- Add comments (MST);
- Refine commit log;
v4:
- s/device/the device/ (Cornelia);
- s/driver/the driver/ (Cornelia);
- Improve commit log and comments (MST);
- Make commit log title more specific;
packed-ring.tex | 23 ++++++++++++++++++++---
1 file changed, 20 insertions(+), 3 deletions(-)
diff --git a/packed-ring.tex b/packed-ring.tex
index f24f49b..0047ecf 100644
--- a/packed-ring.tex
+++ b/packed-ring.tex
@@ -687,16 +687,33 @@ vq->driver_event.flags = RING_EVENT_FLAGS_DISABLE;
for (;;) {
struct pvirtq_desc *d = vq->desc[vq->next_used];
+ /*
+ * Check that
+ * 1. Descriptor has been made available. This check is necessary
+ * if the driver is making new descriptors available in parallel
+ * with this processing of used descriptors (e.g. from another thread).
+ * Note: there are many other ways to check this, e.g.
+ * track the number of outstanding available descriptors or buffers
+ * and check that it's not 0.
+ * 2. Descriptor has been used by the device.
+ */
flags = d->flags;
+ bool avail = flags & VIRTQ_DESC_F_AVAIL;
bool used = flags & VIRTQ_DESC_F_USED;
-
- if (used != vq->used_wrap_count) {
+ if (avail != vq->used_wrap_count || used != vq->used_wrap_count) {
vq->driver_event.flags = RING_EVENT_FLAGS_ENABLE;
memory_barrier();
+ /*
+ * Re-test in case the driver made more descriptors available in
+ * parallel with the used descriptor processing (e.g. from another
+ * thread) and/or the device used more descriptors before the driver
+ * enabled events.
+ */
flags = d->flags;
+ bool avail = flags & VIRTQ_DESC_F_AVAIL;
bool used = flags & VIRTQ_DESC_F_USED;
- if (used != vq->used_wrap_count) {
+ if (avail != vq->used_wrap_count || used != vq->used_wrap_count) {
break;
}
--
2.17.1
---------------------------------------------------------------------
To unsubscribe, e-mail: virtio-dev-unsubscribe@lists.oasis-open.org
For additional commands, e-mail: virtio-dev-help@lists.oasis-open.org
next reply other threads:[~2018-12-05 16:28 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-12-05 16:27 Tiwei Bie [this message]
2018-12-06 12:44 ` [virtio-dev] Re: [PATCH v4] packed-ring: fix used descriptor checking in example code Cornelia Huck
2018-12-07 1:18 ` [virtio-dev] " Tiwei Bie
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20181205162701.14138-1-tiwei.bie@intel.com \
--to=tiwei.bie@intel.com \
--cc=cohuck@redhat.com \
--cc=mst@redhat.com \
--cc=stefanha@redhat.com \
--cc=virtio-dev@lists.oasis-open.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox